On the use of GNSS-inferred crustal strain accumulation in evaluating seismic potential

One of the main applications of GNSS observations is to calculate position time-series, from which velocities are inferred. The horizontal velocities can then be used to calculate crustal strain rates. If the time-series is not dominated by postseismic deformation or nontectonic loading processes, the strain rates typically represent the interseismic strain accumulation in the crust, which is often expected to be accommodated during future earthquakes on regional faults. As such, geodetic strain, or moment, rates can provide important constraints on an area’s seismic potential. We discuss two ways in which geodetic strain rates can be used to evaluate seismic potential with both methods based on the premise that all geodetic deformation is expected to be ultimately accommodated seismically. First, the seismic-to-geodetic moment ratio reveals that past seismic moment typically deviates, and often underestimates, the geodetic moment. Rather than interpreting this as evidence for aseismic deformation, we present confidence limits on the moment ratio as a function of the geodetic potency (i.e., size of the area times length of the earthquake catalog times strain rate) and show that the observed moment ratio is rarely statistically different from parity. Second, we explore the relationship between geodetic strain rates and the number of earthquakes for a certain area and period, forming the basis of geodesy-based earthquake forecasting. This relationship is expected to be linear for all areas that have a similar seismogenic thickness, shear modulus, and tapered power-law magnitude frequency. However, we show that there are relatively fewer events in areas of the highest strain rates, where seismic hazard is the highest, and we summarize possible explanations